weixin_43399179的博客

Overview:In the object detection, bounding boxes are the most catptured our attention. These boxes are used to target the object. Which is 4 dimensional entities defined by left-upper and right-bottom points.There are key components about the object dete

Fully-Convolutional One-Stage Object Detection参考：EECS 498-007/598-005 Assignment 4-1: One-Stage Object Detector论文地址v1论文v2 不同于使用Anchor去确定目标，FCOS是使用全卷积网络通过确定特征图上的每个位置来预测该点距离目标的距离（左上右下 LTRB），如下图：该流程与第一版的网络的不同点在于最后将regression 和centre-ness分支放在一起。该文中使用的是第二版的网络

Assignment2 code: downloaded hereQ1: Multi-Layer Fully Connected Neural NetworksIn this question, we need to implement a fully connected network with an arbitrary number of hidden layers.The main thing to do is implement the initiailzation, forward and

【代码】cs231n-AssignmentNote00-Assignment 1。

​。

𝛾and𝛽.𝛾and𝛽.𝛾and𝛽1. 2.3.

In the previous section, we mostly talked about Neural Networks. CNN is similar to that. Both are made up of neurons with learnable weights and biases. Each neuron receives some inputs and outputs with non-linearity. Still has loss function.The difference